Learning Canonical 3D Object Representation for Fine-Grained Recognition

TL;DR

This paper introduces a framework for fine-grained 3D object recognition from single images, learning shape and appearance variations without ground-truth 3D data, and improving recognition and reconstruction performance.

Contribution

It presents a novel 3D-aware representation learning method that jointly models shape and appearance, enabling viewpoint-invariant recognition without 3D annotations.

Findings

Achieves competitive fine-grained recognition accuracy.

Improves 3D shape reconstruction quality.

Enhances shape deformation learning through boosting.

Abstract

We propose a novel framework for fine-grained object recognition that learns to recover object variation in 3D space from a single image, trained on an image collection without using any ground-truth 3D annotation. We accomplish this by representing an object as a composition of 3D shape and its appearance, while eliminating the effect of camera viewpoint, in a canonical configuration. Unlike conventional methods modeling spatial variation in 2D images only, our method is capable of reconfiguring the appearance feature in a canonical 3D space, thus enabling the subsequent object classifier to be invariant under 3D geometric variation. Our representation also allows us to go beyond existing methods, by incorporating 3D shape variation as an additional cue for object recognition. To learn the model without ground-truth 3D annotation, we deploy a differentiable renderer in an analysis-by-synthesis framework. By incorporating 3D shape and appearance jointly in a deep representation, our method learns the discriminative representation of the object and achieves competitive performance on fine-grained image recognition and vehicle re-identification. We also demonstrate that the performance of 3D shape reconstruction is improved by learning fine-grained shape deformation in a boosting manner.